The Growth in the Social Security Disability Insurance Rolls

The Growth in the Social Security Disability Insurance Rolls

The share of the U.S. population receiving Social Security Disability Insurance (DI) benefits has risen rapidly over the past two decades, from 2.2 percent of adults age 25 to 64 in 1985 to 4.1 percent in 2005. While the share of adults on DI in the U.S. today is still lower than that in most other developed countries, the recent growth of the DI program nonetheless poses significant risks to the finances of the Social Security system. Over the past two decades, the share of total Social Security spending accounted for by DI has risen from 10 percent to 17 percent. In 2005, cash payments to DI beneficiaries topped $85 Billion. DI recipients are also eligible for Medicare two years after the onset of their disability, further boosting the cost of the program.

In The Growth in the Social Security Disability Rolls: A Fiscal Crisis Unfolding (NBER Working Paper 12436), research-ers David Autor and Mark Duggan explore both the causes and consequences of the recent growth in the DI program. They also examine the success of the DI screening process in distinguishing meritorious claims and consider potential reforms to the DI program.

Since its introduction in 1956, the DI program has insured workers against the risk of being unable to work due to disability. To be insured for DI, a person must have worked at least five of the last ten years; currently, more than 80 percent of non-elderly adults in the U.S. meet this criterion. To be awarded DI benefits, individuals must have a medically determinable physical or mental impairment that is expected to result in death or last for at least a year and that prevents them from engaging in "substantial gainful activity." DI applicants go through a medical screening process and may appeal if their initial claim is denied; many applicants do appeal and a significant share of decisions are overturned during the appeals process.

The authors first take up the question of why the disability rolls have grown. The most important factor is the liberalization of the DI screening process that occurred due to a 1984 law. This law directed the Social Security Administration to place more weight on ap-plicants' reported pain and discomfort, relax its screening of mental illness, consider applicants with multiple non-severe ailments, and give more credence to medical evidence provided by the applicant's doctor.

These changes had the effect of both increasing the number of new DI awards and shifting their composition towards claimants with low-mortality disorders. For example, the share of awards for a primary impairment of mental illness rose from 16 percent in 1983 to 25 percent in 2003, while the share for a primary impairment of musculoskeletal disorders (primarily back pain) rose from 13 per-cent in 1983 to 26 percent in 2003.

A second factor is the rising value of DI benefits relative to potential labor market earnings. As the authors explain, this increase is not due to any legislative intent. Rather, the interaction of increasing income inequality and the DI benefit formula means that low-income workers now have a larger share of their pre-disability income replaced at the 90 percent rate and less at the 32 or 15 percent rate. Similarly, there has been a substantial rise in the real value of Medicare received by DI beneficiaries. The authors estimate that the DI replacement rate (including the value of Medicare) for a low-income older male worker rose from 68 percent in 1984 to 86 per-cent in 2004.

By contrast, the authors estimate that the aging of the U.S. population has made only a modest contribution to the growth of the DI program, accounting for 6 percent of the increase. Changes in the health of the population are also deemed to have had a small effect at most.

Next, the authors ask what share of disability recipients are undeserving of their benefit awards ("cheating"). This is a difficult task, as there are no systematic, objective data on the work capacity of DI recipients. Previous studies have established that the labor force participation rate of DI applicants would be 30 to 40 percentage points higher in the absence of the DI program and that this figure has been stable over the years. While this might suggest that today's DI applicants are no more likely to be work-capable than past appli-cants, the authors' evidence suggests otherwise. In previous research, they find that the responsiveness of DI applicants to adverse labor market shocks rose sharply between 1984 and 1998, leading them to conclude that "a growing fraction of discouraged and displaced workers are seeking DI benefits."

Turning to assess the DI screening process, the authors' view is that the process is effectively broken. They note that despite re-peated efforts by the Social Security Administration to improve the efficiency, accuracy, and consistency of the process, it has evolved into an adversarial process relying heavily on appeals and adjudication. In recent years, nearly forty percent of total DI awards were granted during the appeals process, up from 20 percent in the late 1970s. In one recent year, the Social Security Administration was forced to pay nearly a half a billion dollars to claimants' attorneys.

The authors also argue that because the definition of disability adopted in 1984 is quite broad, the DI program often functions in practice as an insurance program for unemployable workers. For example, when 130,000 DI beneficiaries whose primary impairment was drug or alcohol addiction were removed from the DI rolls in 1996, two-thirds of the terminated claimants managed to re-qualify for DI under a different impairment.

Looking ahead, the authors project that the DI program will continue to grow until its rolls include almost 7 percent of the non-elderly adult population, a 70 percent increase over today's enrollment rate. This increase would obviously strain the finances of Social Security and Medicare - the 1.8 percent payroll tax that now covers the DI program would be inadequate, leaving fewer funds available to pay other Social Security benefits, and DI recipients would claim an even bigger share of total Medicare expenditures than the 15 percent they do today.

The authors conclude by considering possible reforms to the DI program. Past experience suggests that efforts to remove people from the DI rolls are not productive, as denied recipients often find their way back on to the program and the public backlash to these removals can aggravate the situation. Tightening the screening process holds more promise, though it seems likely that more deserv-ing applicants would also be denied benefits as a result. Two reforms with more potential are to allow the Social Security Administra-tion to commission independent medical evaluations during the initial screening process and to be represented by an attorney at Ad-ministrative Law Judge hearings. The authors suggest that these reforms would be likely to raise the rejection rate for non-deserving claims while lowering it for deserving claims.

Two more radical options are to make it easier for individuals to obtain health insurance, so that DI is not serving as the insurer of last resort for work-capable people, and to introduce a graduated scale of DI payments, so that those with more severe impairments receive larger benefits. These changes are not without drawbacks, however - the first would cost money and the second might induce more relatively healthy people to apply for DI.

As the authors note, there are no easy fixes for the DI programs. However, "the cost of postponing reforms to DI may eventually come to appear even more daunting than the cost of facing them promptly."

Autor acknowledges financial support from the National Science Foundation (CAREER SES-0239538). Duggan acknowledges financial support from the Alfred P. Sloan Foundation.

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us